The defining feature in neurodegenerative diseases is a deterioration of brain regions that are essential for the proper functioning of cognitive, emotional, behavioural, and motor systems. An increase in the prevalence of diseases that can be treated with medicines is one of the key causes of an overall increase in the average age of the world population, which is getting older. In this regards, non-pharmacological treatments that address cognitive, functional, and neuropsychiatric components are becoming more and more popular. According to recent findings, deep learning performs significantly better than more conventional approaches when it comes to recognising complex high-dimensional information. This is especially true in the realm of computer vision and the analysis of medical images, such as brain images. Because neurodegenerative illnesses have such a negative impact on people quality of life, health care systems all over the world are focusing a lot of their attention and resources on combating them. A growing number of researchers have become interested in the potential applications of deep learning in the diagnosis and classification of diseases. In this paper, we use U-net architecture for the segmentation of brain images and cognitive modelling is applied to this deep learning to improve the ability of learning. The data is split into training and testing samples in order to validate the model. The simulation is conducted for testing the efficiency of the model and the simulation results shows higher accuracy than existing methods

Neurodegenerative Diseases, pharmacological therapy, perplexing structures, U-net architecture